Search free algorithms for DOA estimation of quasi-stationary signals

We propose two search free algorithms for direction-of-arrival (DOA) estimation of quasi-stationary signals. The two algorithms are all based on the Khatri-Rao (KR) subspace. The first algorithm directly estimates the array manifold matrix through joint diagonalization of a set of matrices associated with signal KR subspace. By removing the redundant data in KR signal model, the second algorithm exploits the rotational invariance technique to estimate the DOAs, and it can be used in the case with more sources than sensors. Simulation results are provided to demonstrate the performance of the proposed algorithms.